Resilient Consensus for Discrete-Time Multiagent Systems With a Dynamic Leader and Time Delay: Theory and Experiment

Area of Science:

• Control Theory
• Cybersecurity
• Robotics

Background:

• Cyber-attacks significantly challenge the consensus of multiagent systems (MASs) by compromising individual agents.
• Developing resilient control strategies is crucial for maintaining MAS functionality under adversarial conditions.

Purpose of the Study:

• To address resilient consensus problems in discrete-time MASs under malicious agents, dynamic leaders, and communication delays.
• To design controllers that ensure reliable consensus and bounded errors in MASs.
• To validate the proposed methods through simulations and experiments.

Main Methods:

• Design of a resilient controller for first-order MASs using robust graph concepts.
• Development of a novel graph structure and modified controller to mitigate error bound growth.
• Introduction of an estimator-based control framework for second-order MASs.

Main Results:

• Achieved consensus with ultimately bounded error for first-order MASs.
• Limited error bound growth to a linear rate for larger systems.
• Demonstrated the effectiveness and practicability of proposed methods via simulations and experiments with unmanned vehicles.

Conclusions:

• The proposed control strategies effectively enhance the resilience of discrete-time MASs against cyber-attacks and other disturbances.
• The developed methods are validated for both first- and second-order systems, showing practical applicability in real-world scenarios like autonomous vehicle coordination.